Coronary heart disease (CHD) is the leading cause of mortality in the U.S. Of those who die suddenly from CHD, 50% of men and 64% of women have no previous symptoms. This clinical observation, coupled with our increased appreciation of the interdependence of atherosclerosis, inflammation, and thrombosis, have led many to the conclusion that platelet aggregation and thrombosis is a major factor leading to vasculo-occlusive or atherothrombotic events. Indeed one of the major pharmacological modes for primary and secondary prevention of coronary events is the use of anti-platelet agents, typically aspirin and clopidogrel (Plavix), which have been shown to decrease MI incidence rates by approximately 20-25%. Numerous studies indicate that there is substantial variability in response to anti-platelet agents with up to 30% of subjects considered non-responders to aspirin (1-3), and 25% of subjects considered to be non-responders to clopidogrel (4-6). The ability to predict which individuals will respond to anti-platelet agents and which subjects will be resistant to anti-platelet therapy would have a profound impact on the prevention and treatment of CAD, which would benefit millions of Americans with CAD or who are at risk for CAD. The mechanism underlying inter-individual variability in response to anti-platelet agents is not known. However, some evidence supports an important genetic component. The objective of this proposal is to identify specific gene variants that predict response to clopidogrel therapy. The study will utilize a unique genetics resource that we have established over the past decade. Since 1995, we have worked extensively with the Old Order Amish of Lancaster, PA, a genetically homogeneous closed founder population ideal for genetic studies. Our proposed study will build upon our large ongoing Heredity and Phenotype Intervention (HAPI) Heart Study. Part of the NHLBI-funded PROgram for GENetic Interaction (PROGENI) Network, this study was designed to examine gene-environment interactions in defining risk for CVD. We.are actively recruiting 1,000 Old Order Amish subjects from very large extended pedigrees. All subjects are being extensively characterized with respect to CVD and atherosclerosis and all undergo four short-term interventions, each affecting a different cardiovascular outcome. Most relevant to this PharmGKB proposal, is the aspirin intervention in which 81 mg of aspirin is given for 14 days and platelet aggregation studies are performed before and after the aspirin intervention to quantitate aspirin response. Nearly 300 subjects have already completed the HAPI Heart study. We expect to have recruited all 1,000 subjects by spring 2006. As part of a previous study (Amish Family Calcification Study), we have already completed the genotyping of over 800 STR markers in most of these same 1,000 subjects. These data will be used to conduct a genome-wide linkage analysis for aspirin response as part of the HAPI Heart Study. In this PharmGKB proposal, we propose to follow-up these studies by treating these same subjects with clopidogrel alone followed by the combination of clopidogrel and aspirin, and performing sophisticated measures of platelet aggregation and function, to address the following hypotheses: (1) There will be great inter-individual variation in clopidogrel response, as defined by in vitro platelet function testing, and the prevalence of non-response will be substantial. (2) Clopidogrel response will be largely independent of aspirin response since these agents work through different albeit converging biochemical pathways. (3) Clopidogrel response will be familial, i.e., will be significantly heritable. (4) Specific chromosomal regions and common gene variants will be linked to or associated with clopidogrel response; some of these regions and common gene variants will be distinct from those associated with aspirin response while others (in converging pathways) will be the same. To directly test these hypotheses, the Specific Aims of this proposal are: 1. To perform a two-stage intervention with clopidogrel (300 mg loading dose then 75 mg per day for the next 7 days), followed by clopidogrel 75 mg per day plus aspirin 81 mg per day for 7 days in 800 participants of the Amish HAPI Heart Study. A battery of platelet aggregation studies and other measures of platelet function and inflammatory factors will be performed before and after each intervention. We will then characterize the genetic epidemiology of clopidogrel response by: a. Defining the degree of inter-individual variability in platelet response to clopidogrel and the prevalence of clopidogrel non-responders. b. Defining the relationship between clopidogrel response and aspirin response. c. Estimating the heritability of the response to clopidogrel alone and clopidogrel plus aspirin. d. Determining the clinical correlates of response to clopidogrel (e.g., age, gender, body mass index, inflammatory factors, baseline measures of cardiovascular function, etc.) 2. To identify specific chromosomal loci that influence clopidogrel response by performing genome-wide linkage analysis utilizing a 5 cM density (approximately 800 STR markers) genetic map already available in 1,000 subjects of the Amish HAPI Heart Study. a. To determine whether chromosomal loci linked to clopidogrel response are the same or different from the chromosomal loci linked to aspirin response. 3. To characterize 100 candidate genes for clopidogrel response by: a. Identifying all common sequence variants through DNA sequencing and in silico SNP discovery. b. Validating these SNPs and defining allele frequencies through genotype analysis of approximately equal to 2000 SNPs in the 100 candidate genes in DNA samples from Caucasians and African American subjects. c. Defining linkage disequilibrium between SNPs and haplotype structure of these candidate genes. 4. To identify the SNPs and haplotypes in these candidate genes that influence clopidogrel response by performing family-based association analysis. These analyses will be performed in the first set of 400 HAPI Heart participants with positive associations to be re-run in the second set of 400 participants as a replication. a. To determine whether variants in candidate genes associated with clopidogrel response are the same or different from those associated with aspirin response. ]n summary, this project seeks to determine the genetic underpinnings of the inter-individual variability of clopidogrel (and aspirin) response. It is unique since the study will be performed in very large families (indeed a single 14-generation pedigree) of well-characterized individuals, which will allow us to perform genome-wide linkage analysis as well as more powerful association analysis of candidate genes for drug response.